There is no need to point out here the number and gravity of traffic accidents. For years, a system which is simultaneously reliable, effective and not excessively restrictive, and which allows the risk of collisions between vehicles and, more generally, the number of traffic accidents to be reduced has been sought.
At a very early stage, the use of radar on vehicles for anti-collision purposes was contemplated. One example is the subject described in the journal Electronics, February 1974, entitled "Radar in auto on collision course tightens passengers seat belts". Reading of this document reveals the difficulties of the problem.
These difficulties are fundamental with regard to road traffic as the bodies which are likely to meet one another (in collision) can be driven into extremely varied positions and with extremely varied speed vectors (allowing for both direction and amplitude). Reliable and exhaustive details therefore have to be acquired, but in a sufficiently limited number, on situations likely to cause a collision.
In this particular case, certain stationary objects can be just as important as moving objects
The radar therefore has to recognize the stationary and moving objects which it detects, from their "radar signature". This problem has not yet been solved.
Moreover, the main risk of collision is linked to objects moving at slow speed, transversely to the direction of observation. Now, as the radial speed of such objects is almost zero, it is difficult for radar to distinguish them from stationary objects.
Furthermore, any application of radar is characterised by a "proportion of false alarms". If this proportion is kept very low, dangerous situations will not be indicated to the vehicle driver. Conversely, if it is too high, the driver will lose confidence in the radar which is assisting him.
The search for a satisfactory compromise is therefore extremely awkward in view of the variety of potential collision situations.
Up until now, virtually no commercial solutions have come to light or been highly developed owing to the complexity of the problem.
The object of the present invention is, in particular, to advance the technique toward a solution by proposing means which are much better adapted to the detection of situations likely to cause a collision.